3 safety-motivated departures tracked
Anthropic, founded by former OpenAI researchers as a safety-focused lab, has not been immune to its own safety departures. In February 2026, Head of Safeguards Research Mrinank Sharma resigned warning that 'the world is in peril,' citing a disconnect between Anthropic's stated safety values and the competitive pressures driving actual decisions. Safety researchers Harsh Mehta and Behnam Neyshabur departed alongside him in what became known as the February 2026 safety researcher exodus. Their departures raise a pointed question: if even the lab founded to prioritize safety struggles to maintain that commitment under competitive pressure, what does that signal for the industry?